From animal models to human disease: a genetic approach for personalized medicine in ALS

Abstract

Amyotrophic Lateral Sclerosis (ALS) is the most frequent motor neuron disease in adults. Classical ALS is characterized by the death of upper and lower motor neurons leading to progressive paralysis. Approximately 10 % of ALS patients have familial form of the disease. Numerous different gene mutations have been found in familial cases of ALS, such as mutations in superoxide dismutase 1 (SOD1), TAR DNA-binding protein 43 (TDP-43), fused in sarcoma (FUS), C9ORF72, ubiquilin-2 (UBQLN2), optineurin (OPTN) and others. Multiple animal models were generated to mimic the disease and to test future treatments. However, no animal model fully replicates the spectrum of phenotypes in the human disease and it is difficult to assess how a therapeutic effect in disease models can predict efficacy in humans. Importantly, the genetic and phenotypic heterogeneity of ALS leads to a variety of responses to similar treatment regimens. From this has emerged the concept of personalized medicine (PM), which is a medical scheme that combines study of genetic, environmental and clinical diagnostic testing, including biomarkers, to individualized patient care. In this perspective, we used subgroups of specific ALS-linked gene mutations to go through existing animal models and to provide a comprehensive profile of the differences and similarities between animal models of disease and human disease. Finally, we reviewed application of biomarkers and gene therapies relevant in personalized medicine approach. For instance, this includes viral delivering of antisense oligonucleotide and small interfering RNA in SOD1, TDP-43 and C9orf72 mice models. Promising gene therapies raised possibilities for treating differently the major mutations in familial ALS cases.

Keywords: Amyotrophic lateral sclerosis (ALS); Animal models; Biomarkers; Frontotemporal dementia (FTD); Gene therapy; Mouse; Personalized medicine.

Fig. 1

Timeline of gene discovery and pathogenic mechanisms in ALS. Schematic representation of years of discovery of most important genes implicated in ALS. Mutation in the superoxide dismutase 1 (SOD1) represent approximately 15 % of familial ALS cases (fALS), mutation in C9orf72 represent 35-40 % and both TAR-DNA-binding protein (TARDBP) and Fused in sarcoma (FUS) for 4 % each. Other genes represent less than 1 % each. Protein aggregation and gliosis are pathological hallmark of ALS and have been discovered before the beginning of the illustrated timeline. ER endoplasmic reticulum; NEFH neurofilament heavy; ALS2 alsin; DCTN1 dynactin; PRPH peripherin; SETX senataxin; VAPB vesicle-associated membrane protein-associated protein B; CHMP2B Charged multivesicular body protein 2B; ANG angiogenin; FIG4 phosphoinositide 5-phosphatase; OPTN optineurin; ATXN2 ataxin 2; DAO D-amino acid oxidase; SPG11 spastic paraplegia 11; VCP valosin containing protein; SIGMAR1 sigma non-opioid intracellular receptor 1; TAF15 TATA-box binding protein associated factor 15; UBQLN2 ubiquilin-2; SQSTM1 sequestosome 1; PFN1 profilin-1; HNRNPA1 heterogeneous nuclear ribonucleoprotein A1; ERBB4 erb-2 receptor tyrosine kinase 4; MATR3 matrin 3; TUBA4A tubulin alpha-4a; TBK1 TANK-binding kinase 1

Fig. 2

Clinical findings in Amyotrophic lateral sclerosis (ALS). Signs and symptoms are divided by affected motor neuron. Both upper motor neurons (UMN) and lower motor neurons (LMN) have to be affected for the diagnosis of ALS. Different combination of LMN and UMN signs can be observed. Limb onset is found in around 65 % of patients but most patients will develop signs in both bulbar region and limbs within the course of disease. Up to 50 % of ALS patients may have symptoms of Fronto-Temporal dementia

Fig. 3

Neuropathological findings in human sALS cases and animal models of ALS. Microscopic pictures of neuropathological findings in ALS models. Our previously published TDP-43 and SOD1 mouse models were exploited for illustration of TDP-43 and SOD1 aggregates with permission. a Immunofluorescence microscopy of a hSOD1^G93A mouse spinal cord. The B8H10 antibody was utilized for the specific signal of misfolded SOD1. Pictures were taken at 10x and b 40x magnification for better visualization of aggregates. c-e Double immunofluorescence microscopy of 10 months-old a hTDP-43^G348C mouse spinal cord using hTDP-43 monoclonal antibody and d ubiquitin antibody [138]. Ubiquinated TDP-43 cytoplasmic aggregates can be observed and are typical neuropathological findings in human ALS. f-i Immunofluorescence of 10 months-old hTDP-43^G348C and non-transgenic mice spinal cord. Iba1 antibody f-g and GFAP antibody h-i showed increased microgliosis and astrogliosis in a 10 months-old hTDP-43^G348C mouse. j-l Immunohistochemistry of two human sporadic ALS cases using TDP-43 antibody to illustrated typical neuronal cytoplasmic TDP-43 inclusions in lumbar spinal cord (j), medulla (k) and motor cortex (l). Scale bar = 250 μm (a); 50 μm (b, f-i); 25 μm (c-e); 100 μm (j-l)

Fig. 4

Gene therapy mechanism of action. Schematic representation of possible gene therapy approaches in ALS treatment. All of these approaches can be effective by intra-thecal, intracerebroventricular or peripheral injection of AAV or lentivirus targeting motor neurons or glial cells. a Antisense olinucleotide (ASO) are short synthetic oligonucleotides (15-25 nucleotides) which bind to targeted mRNA. ASO reduces the expression of a specific protein by two main mechanisms. ASO induces the mRNA degradation by endogenous RNase H or blocks the mRNA translation. This is a potential therapeutic avenue in ALS by reducing the protein level of TDP-43, SOD1 of FUS protein level or by targeting of C9orf72 RNA foci. b SiRNAs are double-stranded RNAs which operated through RNA interference pathway. After strand unwinding, one siRNA strand binds argonaute proteins as part of the RNA-induced silencing complex (RISC) and is recruited to a target mRNA which is then cleaved. c Antibodies are another potential therapeutics avenue in ALS [111]. Antibodies can target misfolded proteins and reduce the amount of toxic aggregates. It is suggested that they can reduces the disease propagation between cells. They can also be exploited to block the pathological interaction between proteins by binding to the specific interaction sites. d Gene delivery is another potential therapeutic avenue for loss-of-function mutations. Virus can provide a functional replacement of a missing gene by mRNA or cDNA delivery. This approach was particularly tested in spinal muscular atrophy and revealed great outcomes but is not yet extensively tested in ALS [231]